Crime data from 2020 EDA
Introduction
This analysis aims to explore crime data from Los Angeles using the dataset provided by Data.gov, covering the period starting from 2020. The dataset can be accessed here.
Notes from dataset description:
- This data is transcribed from
original crime reports that are typed on paper and therefore there may
be some inaccuracies within the data.
- Some location fields with
missing data are noted as (0°, 0°).
- Address fields are only
provided to the nearest hundred block in order to maintain privacy.
Loading Dataset
library(tidyverse)
library(dplyr)
library(ggplot2)
library(leaflet)
library(sp)
library(KernSmooth)
library(raster)
library(e1071)
library(data.table)
library(ggcorrplot)
library(plotly)
Sys.setlocale("LC_TIME", "English")## [1] "English_United States.1252"data_main <- read_csv('C:/Users/Przemysław/Desktop/msad/Crime_Data_from_2020_to_Present.csv',show_col_types = FALSE)Initial dataset description
head(data_main){r}{r echo=T, results='hide'} str(data_main)
{r}{r echo=T, results='hide'} summary(data_main)
Missing values investigation
apply(data_main, 2, function(x) any(is.na(x)))## DR_NO Date Rptd DATE OCC TIME OCC AREA
## FALSE FALSE FALSE FALSE FALSE
## AREA NAME Rpt Dist No Part 1-2 Crm Cd Crm Cd Desc
## FALSE FALSE FALSE FALSE FALSE
## Mocodes Vict Age Vict Sex Vict Descent Premis Cd
## TRUE FALSE TRUE TRUE TRUE
## Premis Desc Weapon Used Cd Weapon Desc Status Status Desc
## TRUE TRUE TRUE FALSE FALSE
## Crm Cd 1 Crm Cd 2 Crm Cd 3 Crm Cd 4 LOCATION
## TRUE TRUE TRUE TRUE FALSE
## Cross Street LAT LON
## TRUE FALSE FALSEThere are 10 columns containing N/A values. Moreover, based on the dataset description, the Location columns are converted to (0°, 0°) for entries with missing values. However, there are no zeroed LAT or LNG values in the actual dataset.
sum(data_main$LAT == 0 | data_main$LNG == 0)## [1] 0The percentage impact of the missing values.
missing_percentage <- colMeans(is.na(data_main)) * 100
missing_df <- data.frame(Column = names(missing_percentage), Percent_Missing = missing_percentage)
ggplot(missing_df, aes(x = Column, y = Percent_Missing)) +
geom_bar(stat = "identity", fill = "skyblue", width = 0.5) +
labs(title = "Percentage of Missing Data per Column", x = "Column", y = "Percent Missing") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1))
Based on the missing values percentage, we can drop Crm Cd 3 and Crm Cd 4 as those columns do not contain other values than null.
Values for Mocodes , Vict Descent and Vict Sex could be MNAR (Missing Not At Random) values that are logically associated with each other. The same for Weapon Desc and Weapon Used Cd .
missing_rows_mocodes <- which(is.na(data_main$Mocodes))
missing_rows_descent <- which(is.na(data_main$'Vict Descent'))
missing_rows_sex <- which(is.na(data_main$'Vict Sex'))
#Check if missing values for Vict Sex, Vict Descent and Mocodes are for the same rows.
(identical(missing_rows_mocodes, missing_rows_descent)
&& identical(missing_rows_mocodes, missing_rows_sex)
&& identical(missing_rows_descent, missing_rows_sex))## [1] FALSEThere is no exact matching on Mocodes , Vict Descent and Vict Sex . However Vict Descent and Vict Sex are matching in 99.99% of cases.
length(intersect(missing_rows_descent, missing_rows_sex)) / length(missing_rows_descent) * 100## [1] 99.98901Mocodes and Vict Sex are matching in 94.85%.
length(intersect(missing_rows_mocodes, missing_rows_sex)) / length(missing_rows_mocodes) * 100## [1] 94.85076We can observe that Vict Sex contains hidden missing values represented as 0.
head(data_main$'Vict Age'[is.na(data_main$'Vict Descent')], 100)## [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## [38] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
## [75] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0The occurrence of missing values in Vict Sex and zero values in Vict Age match 99.97% of the time.
length(intersect(missing_rows_sex, which(data_main$'Vict Age' == 0))) / length(missing_rows_sex) * 100## [1] 99.97295Weapon Desc and Weapon Used Cd have exactly matching missing values.
missing_rows_weapon <- which(is.na(data_main$'Weapon Used Cd'))
missing_rows_weapon_desc <- which(is.na(data_main$'Weapon Desc'))
identical(missing_rows_weapon, missing_rows_weapon_desc)## [1] TRUE Summary
- Columns Crm Cd 2, Crm Cd 3,
Crm Cd 4, and Cross Street contain more than 80% missing
values and do not provide valuable information for analysis. These
columns can be dropped in the next steps.
- Columns Weapon
Desc and Weapon Used Cd have matching occurrences of missing
values and are MNAR (Missing Not At Random). This suggests that these
columns have values only if the crime involved the use of a weapon.
- Columns Mocodes, Vict Descent, Vict Sex, and
Vict Age (with 0 as a missing value) have most occurrences
matching, but not exactly. These variables are logically associated with
each other; data is missing if there is an inability to identify a
victim. So, those values are mostly MNAR but there is posibility of MAR
(Missing at random).
Basic dataset processing
# Removing unnecessary columns
data_main <- data_main[, !names(data_main) %in% c("DR_NO",
"Part 1-2",
"AREA", #the same info as AREA NAME
"Crm Cd",
"Premis Cd",
"Status", #the same info as Status Desc
"Mocodes",
"Crm Cd 1",
"Crm Cd 2",
"Crm Cd 2",
"Crm Cd 3",
"Crm Cd 4",
"Cross Street")]Merging together DATE OCC (format MM/DD/YYYY) and TIME OCC (format HHMM),
hour <- as.numeric(substr(data_main$'TIME OCC', 1, 2))
minute <- as.numeric(substr(data_main$'TIME OCC', 3, 4))
date <- as.Date(substr(data_main$'DATE OCC', 1, 10) , format = "%m/%d/%Y")
data_main$datetime_occ <- as.POSIXct(paste(date, hour, minute, sep = " "), format = "%Y-%m-%d %H %M")
data_main = data_main[, !names(data_main) %in% c("DATE OCC", "TIME OCC")]Mapping Vict Descent and Vict Sex . Mapping from LAPD OpenData City of LA.
victim_descent_mapping = c("A" = "Other Asian",
"B" = "Black",
"C" = "Chinese",
"D" = "Cambodian",
"F" = "Filipino",
"G" = "Guamanian",
"H" = "Hispanic/Latin/Mexican",
"I" = "Native",
"J" = "Japanese",
"K" = "Korean",
"L" = "Laotian",
"O" = "Other",
"P" = "Pacific Islander",
"S" = "Samoan",
"U" = "Hawaiian",
"V" = "Vietnamese",
"W" = "White",
"X" = "Unknown",
"Z" = "Asian Indian")
data_main$'Vict Descent' <- factor(data_main$'Vict Descent', levels = names(victim_descent_mapping), labels = victim_descent_mapping)
victim_sex_mapping = c("F" = "Female",
"M" = "Male",
"H" = "Other",
"X" = "Unknown")
data_main$'Vict Sex' <- factor(data_main$'Vict Sex', levels = names(victim_sex_mapping), labels = victim_sex_mapping)Changing 0 in Vict Age into N/A and converting to date in Date Rptd.
data_main$'Vict Age'[data_main$'Vict Age' <= 0] <- NA
data_main$'Date Rptd' <- as.Date(data_main$'Date Rptd', format = "%m/%d/%Y")New columns dow_occ day of week of the crime, hod_occ hour of day, week and year of the crime.
data_main$dow_occ <- weekdays(data_main$datetime_occ)
data_main$hod_occ <- hour(data_main$datetime_occ)
data_main$week <- lubridate::week(data_main$datetime_occ)
data_main$year <- lubridate::year(data_main$datetime_occ)Days between Date Rptd and datetime_occ
data_main$days_diffrence <- as.numeric(data_main$'Date Rptd' - as.Date(data_main$datetime_occ))Basic visual analysis
Visualising volume of crimes by LA area.
barplot(sort(table(data_main$'AREA NAME'), TRUE),
main = "Volume of crimes by LA area ",
ylab = "Count",
col = "skyblue",
border = "black",
las = 2,
cex.names = 0.8,
cex.lab = 1,
cex.main = 1.4,
beside = TRUE
)
Top 25 crime types and locations in LA.
for (val in list(list("crime type", "Crm Cd Desc"), list('location', "Premis Desc"))){
par(mar = c(5, 15, 1, 1))
options(repr.plot.height=100)
barplot(tail(sort(table(data_main[[val[[2]]]])),25),
main = paste("Volume of crimes by", val[[1]], "top 25"),
horiz = TRUE,
xlab = "Count",
col = "skyblue",
border = "black",
las = 2,
cex.names = 0.4,
cex.lab = 0.5,
cex.main = 1,
beside = TRUE,
)
}
Volume of crimes by victim descent.
par(mar = c(5, 10, 1, 1))
barplot(sort(table(subset(data_main, 'Vict Descent' > 0)$'Vict Descent')),
main = "Volume of crimes by victim descent",
horiz = TRUE,
xlab = "Count",
col = "skyblue",
border = "black",
las = 2,
cex.names = 0.9,
cex.lab = 0.5,
cex.main = 1,
beside = TRUE,
)
Volume of crimes by victim descent and victim sex.
df_vict <- na.omit(data_main[, c("Vict Descent", "Vict Sex", "Vict Age")])
df_vict$"Vict Descent" <- as.character(df_vict$"Vict Descent")
names(df_vict) <- c("vict_descent", "vict_sex", "vict_age")
df_vict <- df_vict %>%filter(vict_sex != "Unknown")
ggplot(df_vict, aes(x = factor(vict_descent, levels = names(sort(table(vict_descent), decreasing = TRUE))), fill = vict_sex)) +
geom_bar(position = "dodge", color = "black", alpha = 0.8) +
labs(title = "Volume of crimes with victim descent breakdown by sex", x = "Victim descent", y = "Count") +
theme_minimal() +
theme(
axis.text.x = element_text(angle = 45, hjust = 1),
legend.title = element_blank(),
legend.position = "top"
) 
ggplot(df_vict, aes(x = vict_age, group = vict_sex, color = vict_sex)) +
geom_line(stat = "count") +
labs(title = "Count of victims by age and sex", x = "Age", y = "Count") +
theme_minimal()
ggplot(df_vict, aes(x = vict_sex, y = vict_age, fill = vict_sex)) +
geom_boxplot() +
labs(title = "Victim age by sex", x = "Sex", y = "Age") +
theme_minimal()
Crimes volume in time - hour of day and day of weeks.
dow_hod_df <- data_main %>%
group_by(dow_occ, hod_occ) %>%
summarise(Count = n())%>%
filter(!is.na(dow_occ))
dow_hod_df$dow_occ <- factor(dow_hod_df$dow_occ, levels = c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday"), ordered = TRUE)
ggplot(dow_hod_df, aes(x = dow_occ, y = hod_occ, fill = Count)) +
geom_tile() +
scale_fill_gradient(low = "yellow", high = "red") +
labs(title = "Heatmap of crimes frequency by hour of day and day of week", x = "day of week", y = "hour of day", fill = "frequency") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5)) 
Analysis of number of days between date reported and day of crime
occurrence.
density_days_diffrence <- density(na.omit(as.numeric(data_main$days_diffrence)))
ggplot(data.frame(x = density_days_diffrence$x, y = density_days_diffrence$y), aes(x = x, y = y)) +
geom_line() +
labs(title = "Density Plot of number of days between date reported and day of crime occurrence", x = "days between", y = "Density")
Crime time series visualization with trend.
weekly_counts <- data_main %>%
group_by(year, week) %>%
summarise(count = n())%>%
filter(!(year==2024&week>4))
ggplot(weekly_counts, aes(x = as.Date(paste(year, week, "1", sep = "-"), format = "%Y-%U-%u"), y = count)) +
geom_line() +
geom_smooth(method = "lm", se = FALSE, color = "red", size = 0.3) +
labs(title = "Weekly crimes count with trend", x = "Date", y = "Number of Records") +
scale_x_date(date_breaks = "1 month", date_labels = "%b %Y") +
theme_minimal() +
theme(axis.text.x = element_text(angle = 90, vjust = 0.5)) 
Weapon analysis - top 20 weapons.
#showing action taken
par(mar = c(5, 20, 1, 1))
barplot(tail(sort(table(na.omit(data_main$'Weapon Desc')), FALSE),20),
main = "Crimes with weapon by type",
xlab = "Count",
horiz = TRUE,
col = "skyblue",
border = "black",
las = 1,
cex.names = 0.6,
cex.lab = 1,
cex.main = 1.4,
beside = TRUE
)
Volume of crimes by crime status
par(mar = c(5, 6, 1, 1))
barplot(sort(table(data_main$'Status Desc')),
main = "Volume of crimes by action taken",
horiz = TRUE,
xlab = "Count",
col = "skyblue",
border = "black",
las = 2,
cex.names = 1,
cex.lab = 0.5,
cex.main = 1,
beside = TRUE,
)
Descriptive analysis
Summary for continuous variables.
descriptive_cont <- data_main[, names(data_main) %in% c("Vict Age", "days_diffrence")]
summary(descriptive_cont)## Vict Age days_diffrence
## Min. : 2.00 Min. : 0.00
## 1st Qu.: 28.00 1st Qu.: 0.00
## Median : 37.00 Median : 1.00
## Mean : 39.58 Mean : 11.04
## 3rd Qu.: 50.00 3rd Qu.: 2.00
## Max. :120.00 Max. :1483.00
## NA's :224140 NA's :64Dispersion measurements for continuous variables.
descriptive_cont <- na.omit(descriptive_cont)
summarise(descriptive_cont,
variance_vict_age = var(`Vict Age`),
sd_vict_age = sd(`Vict Age`),
cv_vict_age = sd(`Vict Age`) / mean(`Vict Age`) * 100,
range_vict_age = diff(range(`Vict Age`)),
iqr_vict_age = IQR(`Vict Age`),
variance_days_diffrence = var(`days_diffrence`),
sd_days_diffrence = sd(`days_diffrence`),
cv_days_diffrence = sd(`days_diffrence`) / mean(`days_diffrence`) * 100,
range_days_diffrence = diff(range(`days_diffrence`)),
iqr_days_diffrence = IQR(`days_diffrence`)
)Concentration measurements for continuous variables.
summarise(descriptive_cont,
skewness_vict_age = skewness(`Vict Age`),
kurtosis_vict_age = kurtosis(`Vict Age`),
skewness_days_diffrence = skewness(`days_diffrence`),
kurtosis_days_diffrence = kurtosis(`days_diffrence`)
)Percentage analysis for selected categorical variables.
calculate_percentage_stats <- function(var) {
if (length(unique(data_main[[var]])) > 20) {
data_subset <- subset(data_main, data_main[[var]] %in%
names(sort(table(data_main[[var]]), decreasing = TRUE))[1:20])
return(sort(table(data_subset[[var]]) / nrow(data_main) * 100, TRUE))
}
else {
return(sort(table(data_main[[var]]) / nrow(data_main) * 100, TRUE))
}
}
for (var in c("AREA NAME", "Crm Cd Desc", "Vict Sex", "Vict Descent", "Status Desc", "Weapon Desc", "dow_occ", "Premis Desc")) {
cat(paste("Percentage statistics for", var, ":\n"))
print(calculate_percentage_stats(var))
cat("\n")
}## Percentage statistics for AREA NAME :
##
## Central 77th Street Pacific Southwest Hollywood Southeast
## 6.811366 6.263397 5.824619 5.613965 5.267355 5.053789
## Olympic N Hollywood Newton Wilshire Rampart West LA
## 5.050989 5.005185 4.995554 4.787588 4.723977 4.578726
## Northeast Van Nuys West Valley Harbor Topanga Devonshire
## 4.303566 4.222149 4.202326 4.117438 4.084512 4.074097
## Mission Hollenbeck
## 3.980249 3.731631
##
## Percentage statistics for Crm Cd Desc :
##
## VEHICLE - STOLEN
## 10.736068
## BATTERY - SIMPLE ASSAULT
## 7.953443
## THEFT OF IDENTITY
## 6.198778
## BURGLARY FROM VEHICLE
## 6.172461
## BURGLARY
## 6.121057
## VANDALISM - FELONY ($400 & OVER, ALL CHURCH VANDALISMS)
## 6.063046
## ASSAULT WITH DEADLY WEAPON, AGGRAVATED ASSAULT
## 5.700757
## THEFT PLAIN - PETTY ($950 & UNDER)
## 5.101833
## INTIMATE PARTNER - SIMPLE ASSAULT
## 4.987379
## THEFT FROM MOTOR VEHICLE - PETTY ($950 & UNDER)
## 3.842165
## THEFT FROM MOTOR VEHICLE - GRAND ($950.01 AND OVER)
## 3.578092
## ROBBERY
## 3.421641
## THEFT-GRAND ($950.01 & OVER)EXCPT,GUNS,FOWL,LIVESTK,PROD
## 3.313235
## VANDALISM - MISDEAMEANOR ($399 OR UNDER)
## 2.612399
## SHOPLIFTING - PETTY THEFT ($950 & UNDER)
## 2.526950
## CRIMINAL THREATS - NO WEAPON DISPLAYED
## 2.050543
## BRANDISH WEAPON
## 1.551962
## TRESPASSING
## 1.510414
## INTIMATE PARTNER - AGGRAVATED ASSAULT
## 1.353963
## VIOLATION OF RESTRAINING ORDER
## 1.257428
##
## Percentage statistics for Vict Sex :
##
## Male Female Unknown Other
## 41.07078463 36.60852874 9.06114002 0.01086306
##
## Percentage statistics for Vict Descent :
##
## Hispanic/Latin/Mexican White Black
## 30.488031590 20.252896631 14.101938105
## Unknown Other Other Asian
## 9.978565045 7.902599744 2.195123044
## Korean Filipino Chinese
## 0.543489216 0.427019242 0.395661941
## Japanese Vietnamese Native
## 0.140211931 0.105830890 0.094967825
## Asian Indian Pacific Islander Hawaiian
## 0.050171681 0.026653706 0.019934284
## Cambodian Guamanian Laotian
## 0.008063306 0.007167383 0.006607431
## Samoan
## 0.005375537
##
## Percentage statistics for Status Desc :
##
## Invest Cont Adult Other Adult Arrest Juv Arrest Juv Other UNK
## 8.006504e+01 1.076474e+01 8.669062e+00 3.241001e-01 1.766088e-01 4.479614e-04
##
## Percentage statistics for Weapon Desc :
##
## STRONG-ARM (HANDS, FIST, FEET OR BODILY FORCE)
## 18.5725933
## UNKNOWN WEAPON/OTHER WEAPON
## 3.7358864
## VERBAL THREAT
## 2.5381495
## HAND GUN
## 2.1459593
## SEMI-AUTOMATIC PISTOL
## 0.7772131
## KNIFE WITH BLADE 6INCHES OR LESS
## 0.7321930
## UNKNOWN FIREARM
## 0.7000517
## OTHER KNIFE
## 0.6261381
## MACE/PEPPER SPRAY
## 0.3979017
## VEHICLE
## 0.3459382
## ROCK/THROWN OBJECT
## 0.2917349
## PIPE/METAL PIPE
## 0.2636253
## BOTTLE
## 0.2569059
## FOLDING KNIFE
## 0.2408913
## STICK
## 0.2403313
## CLUB/BAT
## 0.2230848
## KITCHEN KNIFE
## 0.2034865
## AIR PISTOL/REVOLVER/RIFLE/BB GUN
## 0.1981109
## KNIFE WITH BLADE OVER 6 INCHES IN LENGTH
## 0.1800805
## BLUNT INSTRUMENT
## 0.1500671
##
## Percentage statistics for dow_occ :
##
## Friday Saturday Wednesday Monday Thursday Sunday Tuesday
## 15.26563 14.65853 14.20755 14.12142 14.08660 13.92197 13.73114
##
## Percentage statistics for Premis Desc :
##
## STREET
## 25.3506978
## SINGLE FAMILY DWELLING
## 16.8786271
## MULTI-UNIT DWELLING (APARTMENT, DUPLEX, ETC)
## 12.2250917
## PARKING LOT
## 6.9354510
## OTHER BUSINESS
## 4.7408879
## SIDEWALK
## 4.2815034
## VEHICLE, PASSENGER/TRUCK
## 2.9356033
## GARAGE/CARPORT
## 1.9344095
## DRIVEWAY
## 1.6062777
## RESTAURANT/FAST FOOD
## 1.2756822
## DEPARTMENT STORE
## 1.2588836
## PARKING UNDERGROUND/BUILDING
## 0.8483270
## OTHER PREMISE
## 0.7914359
## MARKET
## 0.7614225
## ALLEY
## 0.7134906
## OTHER RESIDENCE
## 0.6897486
## PARK/PLAYGROUND
## 0.6609671
## YARD (RESIDENTIAL/BUSINESS)
## 0.6166189
## GAS STATION
## 0.5739506
## HOTEL
## 0.5698070 TL;DR for visual and descriptive analysis
- The most
frequent LA crimes occur in the central and 77th street
areas, comprising 6.8% and 6.3% of the total, respectively.
- Most
crimes occur on the street, accounting for over 25% of all reported
incidents.
- The most prevalent type of crime is vehicle theft,
representing 10.8% of all reported crimes.
- Victims are
predominantly of Hispanic, Latin, or Mexican descent, constituting over
40% of the total. The distribution of sexes is nearly uniform, with the
mean age around 40 years old.
- Crimes predominantly occur during
nighttime, with a peak on Friday nights, although overall distribution
across days of the week is similar.
- There is a visible increasing
trend in crime volume since 2020.
- The majority of crimes are
reported promptly after they occur.
Relational analysis
Correlogram for a few variables.
data_main <- data_main %>%
mutate(is_investigation_cont = ifelse(data_main$"Status Desc" == "Invest Cont", 1, 0),
is_street = ifelse(data_main$"Premis Desc" == "STREET", 1, 0),
is_vehicle_stolen = ifelse(data_main$"Crm Cd Desc" == "VEHICLE - STOLEN", 1, 0),
is_central = ifelse(data_main$"AREA NAME" == "Central", 1, 0),
is_77_street = ifelse(data_main$"AREA NAME" == "77th Street", 1, 0),
is_male = ifelse(data_main$"Vict Sex" == "Male", 1, 0),
is_weapon = ifelse(is.na(data_main$"Weapon Desc"), 0, 1),
is_theft_of_identity = ifelse(data_main$"Crm Cd Desc" == "THEFT OF IDENTITY", 1, 0))
corr_data <- data_main[, names(data_main) %in% c("Vict Age", "days_diffrence", "is_investigation_cont",
"is_street", "is_vehicle_stolen", "is_weapon",
"is_central", "is_77_street", "is_male", "is_theft_of_identity")]
corr_data <- na.omit(corr_data[sample(nrow(corr_data), 10000), ])
correlation_matrix <- cor(corr_data, use = "complete.obs")
correlation_matrix[is.na(correlation_matrix)] = 0
ggplotly(ggcorrplot(
correlation_matrix, hc.order = TRUE, type = "lower",
outline.col = "white"
))Geospatial analysis
Showing example 100 crimes with description
m <- leaflet() %>%
setView(lng = -118.2437, lat = 34.0522, zoom = 10) %>%
addTiles()
for (i in 1:nrow(tail(data_main, 100))) {
m <- addMarkers(m, lng = data_main$LON[i], lat = data_main$LAT[i], popup = data_main$'Crm Cd Desc'[i])
}
mShowing heatmap of crimes locations.
data_main <- subset(data_main, (data_main$LON != 0) & (data_main$LAT != 0))
kde <- bkde2D(cbind(data_main$LON, data_main$LAT),
bandwidth=c(.0045, .0068), gridsize = c(1e3,1e3))
KernelDensityRaster <- raster(list(x=kde$x1 ,y=kde$x2 ,z = kde$fhat))
KernelDensityRaster@data@values[which(KernelDensityRaster@data@values < 1)] <- NA
palRaster <- colorNumeric("YlOrRd", domain = KernelDensityRaster@data@values, na.color = "transparent")
leaflet() %>%
setView(lng = -118.2437, lat = 34.0522, zoom = 10) %>%
addTiles() %>%
addRasterImage(KernelDensityRaster,
colors = palRaster,
opacity = .8) %>%
addLegend(pal = palRaster,
values = KernelDensityRaster@data@values,
title = "Density of Crimes")Showing example 1000 crimes with weapon
df_weapon <- na.omit(data_main[, c("Weapon Desc", "LAT", "LON")])
m <- leaflet() %>%
setView(lng = -118.2437, lat = 34.0522, zoom = 10) %>%
addTiles()
for (i in 1:nrow(df_weapon %>% sample_n(1000))) {
m <- addMarkers(m, lng = df_weapon$LON[i], lat = df_weapon$LAT[i], popup = df_weapon$'Weapon Desc'[i])
}
m